Internal combustion engines may utilize direct fuel injection, wherein fuel is directly injected in to an engine cylinder to improve mixture preparation and to reduce cylinder charge temperatures. This may be in place of, or in addition to, port fuel injection, wherein fuel is injected into an intake port, upstream of an intake valve of an engine cylinder. An amount of time a direct fuel injector is activated (the direct injection pulse-width) may be a function of fuel pressure supplied to the injector, engine speed and engine load. To leverage the benefits of a direct injection, it may be advantageous to have full control over the pulse width range of the direct fuel injector. This includes a wide range of operating conditions including but not limited to fuel rail pressure, engine speeds and mass fuel flow.
However, the performance of solenoid-controlled direct fuel injectors may have a limitation in their flow characteristics between the ballistic and full lift regions. This region is normally referred to as the transition region of the direct fuel injector. In this region, the flow rate of the fuel injector is inaccurate and unpredictable, causing shot-to-shot as well as part-to-part variability. For example, the direct fuel injector may deliver more or less fuel than desired in the transition region. Further, the variability in the transition region may not show a linear trend, making it difficult to learn and compensate for the variability. The injector variability may cause cylinder torque output imbalance due to the different amount of fuel injected into each cylinder, and may also cause higher tail pipe emissions and reduced fuel economy due to an inability to correctly meter the fuel to be injected into each cylinder. As a result, there may be engine operating regions where the direct fuel injector cannot adequately meet NVH, drivability, and emissions requirements.
Various approaches have been developed to reduce direct injector variability. One example approach is shown by Ranga et al. in US20160153391. Therein, a direct fuel injection is split into multiple injections, one of which has a pulse-width small enough to be delivered in the ballistic region of the direct injector. A transfer function of the injector is learned based on a lambda value and the split ratio. Subsequent direct injection is adjusted based on the learned transfer function.
However, the inventors herein have recognized potential issues with the approach of '391 and other related approaches. As one example, the variability of the injector in the transition region remains unmapped. The various approaches update the transfer function based on a learned injector variability in the ballistic region, wherein the injector pulse-width is smaller than in the transition region. However, there may still be fueling errors for larger injection pulse-widths that are outside the ballistic region but smaller than injection pulse-widths that are inside the lift region. As a result, NVH, drivability, and emissions issues may persist. As an example, in the ballistic region, increasing the electric pulse width to the fuel injector increases the amount of mass delivered. While there may be some variability in the increasing, the variability may be learned and the shape or shift of a slope in the ballistic region can be adjusted to account for the variability. However, in the transition region, an increased pulse width can actually result in the fuel mass injected decreasing. Consequently, it may not be possible to simply shift and reshape the slope of the transition region. The variability is exacerbated due to the slope in the transition region being different for each injector, as well as being significantly different from shot to shot.
In one example, the issues described above may be addressed by a method for an engine comprising: delivering fuel to a cylinder on a combustion cycle as multiple direct injections; and responsive to a fuel mass of one of the multiple direct injections being in a transition region of a direct injector map, updating one or more of a ratio of fuel delivered in each of the multiple direct injections and a number of the multiple injections to move the fuel mass of the one of the multiple direct injections out of the transition region. In this way, direct injector variability in the transition region can be addressed.
As an example, an engine controller may determine an initial fuel injection profile based on engine operating conditions. This may include, for example, a total fuel mass to be delivered via direct injection, a number of direct injections over which to deliver the total fuel mass (e.g., over 2-4 injections), and a split ratio of the multiple direct injections. The split ratio may include a ratio of the portion of the total fuel mass to be delivered via a first of the multiple direct injections relative to the portion of the total fuel mass to be delivered via a second (and third, and so on) of the multiple direct injections. A fuel injector pulse-width to be commanded is then determined based on the split ratio, including based on the fuel mass to be delivered in each of the multiple injections, as well as based on fuel rail pressure. If the pulse-width for any of the injections is determined to be within the transition region of the direct injector, the controller may update the injection profile to operate outside the transition region. In particular, the controller may modify the fuel mass delivered to each of the injections based on their location within the transition region, and their distance from the bordering ballistic and lift regions, thereby updating the split ratio. For example, the fuel mass of an injection close to the ballistic region may be decreased to move the injection from the transition region into the ballistic region. As another example, the fuel mass of an injection close to the lift region may be increased to move the injection from the transition region into the lift region. As such, the fuel mass of all the injections may be adjusted to maintain the total fuel mass.
In still further examples, additionally or optionally, the number of injections of a split injection may be updated. For example, responsive to the pulse-width of at least one of the multiple direct injections being within the transition region, the number of injections may be reduced. By merging the fuel mass of 2 or more injections, the pulse width may be moved into the full lift region (linear region). As another example, responsive to the pulse-width of a single direct injection being within the transition region, the number of injections may be increased. By splitting the fuel mass into 2 or more injections, the pulse width of an injection may be moved into the full lift region while the pulse width of another injection may be moved into the ballistic region. Further still, in engines configured with direct and port injection, a split ratio of the total fuel mass to be delivered via direct injection relative to port injection may be updated. In still further examples, a combination of the above-mentioned approaches may be selected. The selection may be based on engine operating conditions such as engine speed and NVH constraints.
In this way, direct injector variability is reduced. The technical effect of adjusting a direct injection fuel mass based on a location of the pulse width of the injection on a map of direct injector operating regions is that the direct fuel injector may not be operated at pulse widths where non-linear fuel injector behavior occurs. At the same time, a total fuel mass may be maintained. As a result of operating outside the transition region of the direct injector, engine air-fuel ratio and torque errors may be reduced. Further, the approach may reduce engine emissions and NVH issues. Overall, drivability is improved.
It should be understood that the summary above is provided to introduce in simplified form a selection of concepts that are further described in the detailed description. It is not meant to identify key or essential features of the claimed subject matter, the scope of which is defined uniquely by the claims that follow the detailed description. Furthermore, the claimed subject matter is not limited to implementations that solve any disadvantages noted above or in any part of this disclosure.